gf2121 commented on issue #15820: URL: https://github.com/apache/lucene/issues/15820#issuecomment-4065638581
> TrieReader replaces in-memory FST with on-disk trie navigation. The FST was loaded into Java heap at segment open time — navigation was pure CPU (BitTable.isBitSet). The TrieReader reads from memory-mapped files, adding I/O indirection. Actually Lucene has been loading FST off-heap by default since [LUCENE-9257](https://issues.apache.org/jira/browse/LUCENE-9257) so I don't think the issue is related to heap memory usage. According to your profile, the performance bottleneck likely stems from the difference between the new Panama-based Foreign Memory mmap and the old ByteBuffer mmap. Maybe testing with TrieReader + ByteBuffer mmap could help us confirm the root cause of the regression. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
